Metabolisme, consumd’energia i origens del nostrebackground genètic per a l’ús

de la dieta com a teràpiaSandra Acosta, PhD

100 anys de l’Origen de les espècies de Darwin

La dieta és un factor de pressió adaptativa molt important

Neodarwinisme o la mutació coma motor de l’evolució

Els humans som primats

I si som tan iguals als altres primats, per què som tan diferents?

Evolució humana i la dietaHunter-gatherers

Evolució humana i la dieta

La revolució neolítica

Evolució humana i la dieta

a trade-off between early onset of reproductionand cessation of growth (14, 15). Perturbations inthe GH1-IGF1 pathway have been implicated toplay a role in short stature in RFHG populationsin central Africa and southeast Asia (16).In central African RFHG populations where

admixture with neighboring populations is com-mon, short stature is significantly correlated withancestry and is highly heritable (14). Central AfricanRFHG–specific genomic adaptationswere identifiedthrough comparisons with other African popu-lations by using methods such as locus-specificbranch length (LSBL) (17) and XP-EHH (3) tests,which have high power to detect population-specific positive selection (Fig. 3B). These testsidentified a 15-Mb region on chromosome 3 thatshows signatures of strong positive selection (14).Several genes in this region are associated withshort stature in Pygmies, includingDOCK3, whichis associatedwithheight variation innon-Africans,andCISH, which is important in immune responsebut also inhibits human growth hormone receptoractivity, indicating that short stature could be aproduct of selection acting on pleiotropic loci(14). A subsequent study discovered a ~200-kbhaplotype containing HESX1 at high frequencyin central African RFHGs but low frequency inother African populations; HESX1 is involved inthe development of the anterior pituitary, where

growth hormone is produced (18). This haplotypewas not tagged in previous genotyping arrays,demonstrating the importance of including eth-nically diverse populations in whole-genomesequencing studies (18). However, RFHG andneighboring agriculturalist populations inUgandaexhibit a distinct set of loci associated with shortstature, which raises the possibility of convergentevolutionof this trait across theRFHGs inAfrica (15).The studies discussed above suggest that short

stature in central African RFHG is mostly drivenby strong selection affecting a relatively small num-ber of loci ofmoderate to strong effect. In contrast,hundreds of loci are associated with Europeanheight variation (19, 20). Detecting selection on apolygenic trait such as height is difficult becausethe causative variants can be ancient polymor-phisms each of relatively small effect. Several re-centmethodsuseGWAS to identify trait-associatedvariants then test for allele frequency shifts be-tween populations [either unweighted (20) orweighted by effect size (21)] greater than expectedfrom neutral drift. Polygenic selection tests usingGWASdata (Fig. 2C)have suggestedweak selectioninfluencing average height differences amongEuropean populations, with Northern Europeansgenerally being taller than Southern Europeans(20, 21). A recent aDNA analysis of NeolithicEuropean populations suggests that the North-

South European height gradient may reflect selec-tion for shorter height in EarlyNeolithicmigrantsinto southern Europe and admixture of tallersteppe populations with northern Europeans (11).Although analysis of aDNA holds great promisefor revealing human phenotypic history, this ap-proach faces challenges for studies of indigenouspopulations such as RFHG because DNA is notwell-preserved in tropical climates, and large-scaleGWAS are not available.

Adaptation to endemic pathogens

Pathogenic environments are an important driverof local adaptation in humans (22), and nowhereis the challenge of survival from pathogens greaterthan in tropical rainforests. In particular,malaria, amosquito-borne protozoan parasitic infection, is amajor cause of mortality in sub-Saharan Africa.The most lethal malarial species, Plasmodiumfalciparum, causes>1million childdeaths inAfricaeach year (23). Tropical populations have longbeenknown to have genetic variants that confer resist-ance to malaria, including the sickle cell, a+- andb+-thalassemia–causing alleles at the hemoglo-bin loci, as well as variants at ABO, GYPA, GYPB,GYPE, and G6PD [reviewed in (23)]. These var-iants are likely adaptive, as initially evidenced bythe strong correlation between allele frequencyand prevalence of malaria infection. However,

SCIENCE sciencemag.org 7 OCTOBER 2016 • VOL 354 ISSUE 6308 55

Lactase persistence

Skin pigmentation

Height

High altitude

Arctic environment High-fat diet

Toxic arsenic-rich environments

Thick hair

MalariaTrypanosome resistance

MCM6

MCM6

Polygenic

FADS loci

Starchy food

Increased BMI

CPT1ALRP5THADAPRKG1

EGLN1EPAS1

SLC24A5SLC45A2TYRMC1R

CISHDOCK3STAT5HESX1POU1F1

VAV3ARNT2THRB

APOL1

G6PDHBBGYPAGYPBGYPC

EDAR1

AMY1

EGLN1

AS3MT

CREBRF

Key

Fig. 1. Examples of human local adaptations, each labeled by the phenotype and/or selection pressure, and the genetic loci under selection. [Adaptedfrom (13)]

on Novem

ber 24, 2019

http://science.sciencemag.org/

Dow

nloaded from

Com es seleccionen els tretsfenotípics complexes?

several malaria-protective variants also causecommon Mendelian diseases in hemi- or homo-zygotes [such as sickle cell anemia, G6PD defi-ciency, and thalassemia (23)] and aremaintainedthrough balancing selection.Another example of adaptation to pathogens

resulting in high frequencies of genetic variantsassociated with disease is at the APOL1 locus.The “G1” and “G2” genetic variants in the last exonof APOL1 are associated with chronic kidney dis-ease, which is disproportionately common amongindividuals of African descent (24). These variantsalso confer resistance to human African trypano-somes (causing “African sleeping sickness”) bymodifying the protein to lyse Trypanosoma bruceirhodesiense, a parasitic protozoa transmitted bythe tsetse fly. The region flanking the G1 andG2 variants appear to be under recent positiveselection, as indicated by extended haplotypehomozygosity (25). G1 andG2 are common in somewesternAfrican populations, but T.b. rhodesienseis currently common only in eastern and south-ern Africa, which raises questions about the evo-lutionary history of the APOL1 locus and thepleiotropic effects of G1 and G2 (25).

Adaptation to high altitude

Living at high altitude (>2500 m above sea level)in regions such as the Tibetan Plateau, the AndeanAltiplano, and the Semien plateau of Ethiopia canbe deadly because of an insufficient supply of oxy-gen to vital organs (hypoxia).However, populationsliving in these high-altitude regions for thousandsof years have adapted and thrived, with varyingphysiological adaptations to hypoxic environments[reviewed in (26)].Recent genome-wide scans of selection (such as

LSBL, PBS, iHS, and XP-EHH) (Fig. 3B) uncoveredgenetic adaptations to high altitude by comparingAndean, Ethiopian, and Tibetan genomes withlowland populations with similar genetic ances-try [reviewed in (27)]. Signatures of positive se-lection were found repeatedly at genes involvedin the hypoxia-inducible factor (HIF) pathway[reviewed in (28)] but on different haplotypebackgrounds (for example, EGLN1 in Andeanand Tibetan populations) owing to convergentevolution (29). One of the strongest signals ofselection in the Tibetan populations is at EPAS1,a transcription factor influencing theHIFpathway.Sequence analysis suggests that the selected Tibetanhaplotype may have originated from introgressionof genomic DNA from Denisovans (Fig. 2D) (30).

Adaptation to toxic environments

Arsenic is acutely toxic to humans but is naturallypresent at high levels in groundwater across theglobe (31). San Antonio de los Cobres (SAC),Argentina, is one such locale with high levels ofarsenic yet has been settled by human popula-tions for the past ~11,000 years. Arsenic metab-olism involves methylating inorganic arsenic tomonomethylarsonic acid (MMA) and subsequentlyto dimethylarsinic acid (DMA), which is less toxic.The fraction of arsenic metabolites (%DMA/%MMA) inurine indicates the efficiency of arsenicmetabolism. Individuals in SAC showparticularly

56 7 OCTOBER 2016 • VOL 354 ISSUE 6308 sciencemag.org SCIENCE

A Hard sweep

B Soft sweep

C Polygenic adaptation

D Adaptive introgression

i) Selection on di!erent de novo mutations

ii) Selection on standing variations

Fig. 2. Genomic signatures of local adaptation. (A) A hard sweep. (B) A soft sweep acting on (i) multipledenovobeneficialmutations and (ii) standing variation. (C) Polygenic adaptation. (D) Adaptive introgression.Each horizontal bar represents a haplotype (a sequence of genetic variants on the same chromosome).Theorange segment is graded to indicate the strength of linkage disequilibrium between beneficial (stars) andneutral (dots) variants.

GENES AND ENVIRONMENT

on Novem

ber 24, 2019

http://science.sciencemag.org/

Dow

nloaded from

Trets poligenics

54 7 OCTOBER 2016 • VOL 354 ISSUE 6308 sciencemag.org SCIENCE

REVIEW

Going global by adapting local: Areview of recent human adaptationShaohua Fan,1* Matthew E. B. Hansen,1* Yancy Lo,1,2* Sarah A. Tishkoff1,3†

The spread of modern humans across the globe has led to genetic adaptations to diverselocal environments. Recent developments in genomic technologies, statistical analyses,and expanded sampled populations have led to improved identification and fine-mappingof genetic variants associated with adaptations to regional living conditions anddietary practices. Ongoing efforts in sequencing genomes of indigenous populations,accompanied by the growing availability of “-omics” and ancient DNA data, promises a newera in our understanding of recent human evolution and the origins of variable traits anddisease risks.

Modernhumans originated~200,000 yearsago inAfrica. Over the past 100,000 years,humans spread across the globe into avariety of habitats, from tropical to arc-tic, from high altitudes to lowlands and

even to toxic environments. After humans mi-grated out of Africa, they encountered andinterbred with archaic populations such asNeandertals and Denisovans, resulting in in-trogression of archaic genomes into non-Africanmodern human genomes (~1 to 6% of moderngenomes) (1). Introgression within Africa alsolikely occurred (2) but is more challenging toquantify because ancient DNA (aDNA) does notpreserve well in that region, and no archaicAfrican genomes are currently available. Withinthe past 10,000 years, most human populationshave transitioned from a hunting-gathering life-style to practicing agriculture and pastoralism,resulting in rapid population growth, increasedpopulation densities, and an increase in infec-tious diseases. The selection pressures for adapt-ing to local environments and new diets haveresulted in population- or region-specific geneticvariants that influence variable phenotypes (suchas height, innate immune response, lactose tol-erance, fatty acid metabolic efficiency, and he-moglobin levels).Establishing a complete picture of local hu-

man adaptation can be challenging because itinvolves identifying the genomic regions underselection, the phenotypes that selection is actingupon, and ideally, the external conditions drivingthe selection. Populations that have adapted toenvironments that severely challenge survivalprovide well-characterized cases for local adap-tation. Here, we review several recent examplesthat illustrate the use of emerging data, such as

aDNAandgenome-wideassociation studies (GWAS),and the impact these adaptations have for diseaserisk (Fig. 1).

Adaptation to dairy consumption

The advent of cattle domestication in the MiddleEast and North Africa, ~10,000 years ago, lead tostrong selection pressure for the ability to drinkmilk as adults. Variants near the LCT locus—coding for the lactase enzyme that metabolizeslactose, the main carbohydrate in milk—showsome of the strongest signals of selection in thehumangenome (3). Inmostmammals and inmosthumans, the level of the lactase enzyme decreasesafterweaning [lactasenonpersistence (LNP)].How-ever, many populations that have traditionallypracticed dairyingmaintain high levels of lactaseinto adulthood [lactase persistence (LP)].A genetic variant associatedwith LP in Europeans

was mapped to intron 13 of MCM6 upstream ofLCT (4). Additional variants located within 100base pairs (bp) of the European variant were iden-tified in African populations (5). The LCT regionwas repeatedly reported as a target of a recentstrong selective sweep (Fig. 2A) in Europeansand Africans, on the basis of numerous statisticaltests including allele frequency comparisons be-tween global populations [fixation index (FST)analyses] and extended haplotype homozygositytests within [extended haplotype homozygosity(EHH) and integrated haplotype score (iHS)] andacross [cross-population EHH (XP-EHH)] pop-ulations (Fig. 3) (3, 6, 7). Indeed, the homozygosityof Africanhaplotypes containing theLP-associatedvariants extends on average nearly 2 Mb, but only~2000 bp in ancestral haplotypes (5, 8). In vitrostudies showed that these derived alleles enhancethe expression of LCT (5, 9). The European LP-associated variant is estimated to be ~9000 yearsold, whereas the most common East African LPvariant is ~5000 years old, which is consistentwith archeological evidence for cattle domestica-tion in the Middle East and east Africa (5). Se-quencing of aDNA indicates that the EuropeanLP-associated allele was absent in early NeolithicCentral Europeans and was at low frequency in

late Neolithic Europeans (10), suggesting that LPspread recently (within the past ~4000 years)across Europe (11).The genetic adaptations resulting in LP are

examples of convergent evolution in modern hu-mans (causative genetic variants arose indepen-dently in geographically diverse populations owingto strong selective pressure for an adaptive phe-notype). However, the identified LP-associatedvariants do not entirely explain the LP pheno-type, particularly in western Africans and somecentral and southern Asian pastoralist popula-tions. For example, the Fulani populations fromNigeria and Cameroon have the European variantat moderate frequency but lack the eastern AfricanLP-associated variants and have a distinct haplo-type with extended homozygosity near LCT, sug-gesting the presence of additional unknownfunctional variants (8). Furthermore, an outstand-ing question is the role that the human gut mic-robiome plays in LP.

Adaptation to an arctic environment

The Inuit populations in Alaska, Canada, andGreenland have adapted to a cold and dark Arcticenvironment and a marine diet rich in omega-3polyunsaturated fatty acids. A recent study com-pared genomic diversity in Greenlandic Inuitswith Europeans andChinese using the populationbranch statistic (PBS) (Fig. 3B, iii) (12) and foundthat themost differentiated region encompasses agene cluster that codes for fatty acid desaturaseenzymes (FADS), which are important modula-tors of fatty acid composition. Two variants inthe FADS region were significantly associatedwith short stature in the Inuits, possibly becauseof the influence of fatty acid composition on growthhormone regulation (12). These variants were alsoassociatedwithheight in a larger studyofEuropeans,but the variants were present at low frequencies andwould have been hard to discover without studyingthe Inuit population, demonstrating why studiesof indigenous populations can be informative foridentifying variants of functional importanceacross ethnic groups (13).

Adaptation to tropical rainforests

Tropical rainforests are some of the harshest en-vironments in the world, characterized by hightemperature and humidity as well as the prev-alence of parasites and other pathogens. Individ-uals living in tropical environments often havevery short life spans, which directly affects repro-ductive success and, hence, can act as a selectivepressure.

Short stature

One distinctive phenotype thought to be adapt-ive to a tropical rainforest environment is shortstature (commonly referred to as a “pygmy” phe-notype), which is defined as an average height of<150 cm in adult males. The short stature trait isan example of convergent evolution in rainforesthunter-gatherer (RFHG) populations across Africa,Asia, and South America. Selection for small bodysize may be due to limited food resources, re-sistance to heat stress, immune response, and/or

1Department of Genetics, University of Pennsylvania,Philadelphia, PA 19104, USA. 2Institute for BiomedicalInformatics, University of Pennsylvania, Philadelphia, PA19104, USA. 3Department of Biology, University ofPennsylvania, Philadelphia, PA 19104, USA.*These authors contributed equally to this work. †Correspondingauthor. Email: tishkoff@mail.med.upenn.edu

GENES AND ENVIRONMENT

on Novem

ber 24, 2019

http://science.sciencemag.org/

Dow

nloaded from

pseudogenes14. However, there is currently little evidence to suggestthat the addition of novel genes is a major mechanism in humanbrain evolution13.Recent studies suggest that human brain evolution is associated

with changes in gene expression specifically within the brain asopposed to other tissues such as liver. A few studies suggest more-accelerated gene expression changes in the brain along the humanlineage compared with the chimpanzee lineage15. Although thestudies differ in design and principal conclusions, they share supportfor an increase in expression level in a subset of brain-expressed genesin the lineage leading to humans16,17.

There is also accumulating evidence that some neural genesunderwent important changes in their coding sequence over thecourse of recent brain evolution, although the proportion of neuralgenes that were targets of positive selection is still in debate. Genesstrongly influenced by natural selection can be identified by compar-ing DNA changes that occur in different, closely related species, forexample in different primate species. Synonymous DNA substi-tutions do not alter the amino acid sequence because they occur atdegenerate sites in the codon (such as a CGT to CGG change, as bothcodons encode arginine). Because synonymous changes do not alterthe biochemical properties of the encoded protein, they are usually

Figure 1 | Differences in cerebral cortical size are associated withdifferences in the cerebral cortex circuit diagram. The top panel shows sideviews of the brain of a rodent (mouse), a chimpanzee and a human to showrelative sizes. The middle panel shows a cross-section of a human andchimpanzee brain, with the cellular composition of the cortex illustrated inthe bottom panel (adapted from ref. 5). The cerebral cortex derives from twodevelopmental cell populations: the primordial plexiform layer (PPL) andthe cortical plate (CP). The primordial plexiform layer seems to behomologous to simple cortical structures in Amphibia and Reptilia, andappears first temporally duringmammalian brain development. The corticalplate develops as a second population that splits the primordial plexiform

layer into two layers (layer I at the top and the subplate (SP) at the bottom;numbering follows the scheme of ref. 31). Cortical-plate-derived corticallayers are added developmentally from deeper first (VI, V) to moresuperficial (III, II) last. Cortical-plate-derived cortical layers areprogressively elaborated in mammals with larger brains (for example,insectivores have a single layer II/III/IV that is progressively subdivided intoII, III, IV, then IIa, IIb, and so on), so that humans have a larger proportionof these late-derived neurons, which project locally or elsewhere within thecortex. Images from the top and middle panels are from the ComparativeBrain Atlas (http://www.brainmuseum.org).

NATURE|Vol 437|1 September 2005 PROGRESS

65© 2005 Nature Publishing Group

pseudogenes14. However, there is currently little evidence to suggestthat the addition of novel genes is a major mechanism in humanbrain evolution13.Recent studies suggest that human brain evolution is associated

with changes in gene expression specifically within the brain asopposed to other tissues such as liver. A few studies suggest more-accelerated gene expression changes in the brain along the humanlineage compared with the chimpanzee lineage15. Although thestudies differ in design and principal conclusions, they share supportfor an increase in expression level in a subset of brain-expressed genesin the lineage leading to humans16,17.

There is also accumulating evidence that some neural genesunderwent important changes in their coding sequence over thecourse of recent brain evolution, although the proportion of neuralgenes that were targets of positive selection is still in debate. Genesstrongly influenced by natural selection can be identified by compar-ing DNA changes that occur in different, closely related species, forexample in different primate species. Synonymous DNA substi-tutions do not alter the amino acid sequence because they occur atdegenerate sites in the codon (such as a CGT to CGG change, as bothcodons encode arginine). Because synonymous changes do not alterthe biochemical properties of the encoded protein, they are usually

Figure 1 | Differences in cerebral cortical size are associated withdifferences in the cerebral cortex circuit diagram. The top panel shows sideviews of the brain of a rodent (mouse), a chimpanzee and a human to showrelative sizes. The middle panel shows a cross-section of a human andchimpanzee brain, with the cellular composition of the cortex illustrated inthe bottom panel (adapted from ref. 5). The cerebral cortex derives from twodevelopmental cell populations: the primordial plexiform layer (PPL) andthe cortical plate (CP). The primordial plexiform layer seems to behomologous to simple cortical structures in Amphibia and Reptilia, andappears first temporally duringmammalian brain development. The corticalplate develops as a second population that splits the primordial plexiform

layer into two layers (layer I at the top and the subplate (SP) at the bottom;numbering follows the scheme of ref. 31). Cortical-plate-derived corticallayers are added developmentally from deeper first (VI, V) to moresuperficial (III, II) last. Cortical-plate-derived cortical layers areprogressively elaborated in mammals with larger brains (for example,insectivores have a single layer II/III/IV that is progressively subdivided intoII, III, IV, then IIa, IIb, and so on), so that humans have a larger proportionof these late-derived neurons, which project locally or elsewhere within thecortex. Images from the top and middle panels are from the ComparativeBrain Atlas (http://www.brainmuseum.org).

NATURE|Vol 437|1 September 2005 PROGRESS

65© 2005 Nature Publishing Group

pseudogenes14. However, there is currently little evidence to suggestthat the addition of novel genes is a major mechanism in humanbrain evolution13.Recent studies suggest that human brain evolution is associated

with changes in gene expression specifically within the brain asopposed to other tissues such as liver. A few studies suggest more-accelerated gene expression changes in the brain along the humanlineage compared with the chimpanzee lineage15. Although thestudies differ in design and principal conclusions, they share supportfor an increase in expression level in a subset of brain-expressed genesin the lineage leading to humans16,17.

There is also accumulating evidence that some neural genesunderwent important changes in their coding sequence over thecourse of recent brain evolution, although the proportion of neuralgenes that were targets of positive selection is still in debate. Genesstrongly influenced by natural selection can be identified by compar-ing DNA changes that occur in different, closely related species, forexample in different primate species. Synonymous DNA substi-tutions do not alter the amino acid sequence because they occur atdegenerate sites in the codon (such as a CGT to CGG change, as bothcodons encode arginine). Because synonymous changes do not alterthe biochemical properties of the encoded protein, they are usually

Figure 1 | Differences in cerebral cortical size are associated withdifferences in the cerebral cortex circuit diagram. The top panel shows sideviews of the brain of a rodent (mouse), a chimpanzee and a human to showrelative sizes. The middle panel shows a cross-section of a human andchimpanzee brain, with the cellular composition of the cortex illustrated inthe bottom panel (adapted from ref. 5). The cerebral cortex derives from twodevelopmental cell populations: the primordial plexiform layer (PPL) andthe cortical plate (CP). The primordial plexiform layer seems to behomologous to simple cortical structures in Amphibia and Reptilia, andappears first temporally duringmammalian brain development. The corticalplate develops as a second population that splits the primordial plexiform

layer into two layers (layer I at the top and the subplate (SP) at the bottom;numbering follows the scheme of ref. 31). Cortical-plate-derived corticallayers are added developmentally from deeper first (VI, V) to moresuperficial (III, II) last. Cortical-plate-derived cortical layers areprogressively elaborated in mammals with larger brains (for example,insectivores have a single layer II/III/IV that is progressively subdivided intoII, III, IV, then IIa, IIb, and so on), so that humans have a larger proportionof these late-derived neurons, which project locally or elsewhere within thecortex. Images from the top and middle panels are from the ComparativeBrain Atlas (http://www.brainmuseum.org).

NATURE|Vol 437|1 September 2005 PROGRESS

65© 2005 Nature Publishing Group

Mouse

Chimp

Human

NeuronsSynapsesGliaBlood vessels

Com ha evolucionat el nostre cervell?

El nostre cervell és l’organ que més energiaconsumeix del nostre cos

20% en adults i més d’un 40% en infants

AstrocytesBlood supply

Source of brain energy

Neurons

A

Aq4 tdTOMATO

Knock-inmESC astrocytereporterline

Neomycin Aq4

demonstrate the possibility of using PSC-based in vitro models forstudying synaptic development in the human, although much workwill be necessary to relate in vitro findings to in vivo situations. Inthis context, two lines of investigations should probably bedeveloped further to assess the connectivity patterns of humanneurons in a more physiologically relevant way. On the one hand, itwill be interesting to use 3Dmodels of corticogenesis, or other brainstructures, to dissect precise patterns of synaptic connectivity(Kadoshima et al., 2013; Lancaster et al., 2013). On the other hand,the transplantation of human PSC-derived cells into recipient mice,which enables human neuron development and connectivity to befollowed for up to one year (Espuny-Camacho et al., 2013), couldconstitute a promising tool to study the impact of genes orenvironment on human neuron connectivity in an in vivo contextthat is experimentally tractable.

Conclusions, challenges and future perspectivesWhile it has become trivial to discuss the potential of PSCtechnology for translational purposes, such as disease modeling andcell therapy, their potential usefulness as a tool in basicdevelopmental biology had remained less clear. From the datadiscussed here, PSC-derivedmodels of neural development have thepotential to do much more than just extend the findings previouslyobtained in animal models to the human setting. In particular, PSCmodels have revealed, in sometimes striking ways, the extent to

which cell diversity and spatial and temporal patterning can berobustly recapitulated in simplistic in vitro settings, emphasizing theimportance of intrinsic self-organization during development. Eventhough this could be considered trivial (after all, the embryo itself isobviously capable of self-organization), the versatility andaccessibility of PSC models makes them ideal to study theunderlying mechanisms, in particular for those structures that arespecific to organisms developing in utero. This should constitute auseful source of inspiration to design new experiments in vitro, aswell in in vivo models, to understand the underlying mechanisms.

Another striking observation is that PSC models recapitulatefaithfully the timing of developmental events, and in a species-specific way. PSCs thereby constitute promising tools to uncoverand dissect the links between development and evolution, not onlyby comparing human and mouse PSC-derived systems, but also byimplementing reprogramming of somatic cells of other mammalianspecies in which experimental manipulation in embryos is difficult,such as non-human primates (Wunderlich et al., 2014). This mightshed further light on the evolutionary conservation and divergenceof various other aspects of brain development, such as the gyrationpatterns of the cerebral cortex that have been independently acquiredin multiple mammalian lineages (Lui et al., 2011).

From a molecular perspective, a number of genes have beenidentified recently, mostly through transcriptomics/genomicsapproaches, that may be involved in human-specific features of brain

1 month

B Mouse

ACortical

differentiation

Mouse or human PSCs PSC-derivedcortical cell

Dissociation Transplantation

Cortex

3 weeks 2 months 6 months 9 months

C Human

Fig. 5. Modeling neuronal maturation usingxenotransplantation. Cortical pyramidal neuronsderived from mouse or human PSCs displayspecies-specific maturation followingtransplantation in the mouse neonatal cortex (A).Whereas a mature-like pattern can be observed1 month post-transplantation with mouse ESC-derived cortical cells (B), human cells reach a similarstage 9 months post-transplantation (C). Scalebars: 20 µm. Images are adapted from Espuny-Camacho et al. (2013) and Gaspard et al. (2008).

3146

REVIEW Development (2015) 142, 3138-3150 doi:10.1242/dev.120568

DEVELOPM

ENT

demonstrate the possibility of using PSC-based in vitro models forstudying synaptic development in the human, although much workwill be necessary to relate in vitro findings to in vivo situations. Inthis context, two lines of investigations should probably bedeveloped further to assess the connectivity patterns of humanneurons in a more physiologically relevant way. On the one hand, itwill be interesting to use 3Dmodels of corticogenesis, or other brainstructures, to dissect precise patterns of synaptic connectivity(Kadoshima et al., 2013; Lancaster et al., 2013). On the other hand,the transplantation of human PSC-derived cells into recipient mice,which enables human neuron development and connectivity to befollowed for up to one year (Espuny-Camacho et al., 2013), couldconstitute a promising tool to study the impact of genes orenvironment on human neuron connectivity in an in vivo contextthat is experimentally tractable.

Conclusions, challenges and future perspectivesWhile it has become trivial to discuss the potential of PSCtechnology for translational purposes, such as disease modeling andcell therapy, their potential usefulness as a tool in basicdevelopmental biology had remained less clear. From the datadiscussed here, PSC-derivedmodels of neural development have thepotential to do much more than just extend the findings previouslyobtained in animal models to the human setting. In particular, PSCmodels have revealed, in sometimes striking ways, the extent to

which cell diversity and spatial and temporal patterning can berobustly recapitulated in simplistic in vitro settings, emphasizing theimportance of intrinsic self-organization during development. Eventhough this could be considered trivial (after all, the embryo itself isobviously capable of self-organization), the versatility andaccessibility of PSC models makes them ideal to study theunderlying mechanisms, in particular for those structures that arespecific to organisms developing in utero. This should constitute auseful source of inspiration to design new experiments in vitro, aswell in in vivo models, to understand the underlying mechanisms.

Another striking observation is that PSC models recapitulatefaithfully the timing of developmental events, and in a species-specific way. PSCs thereby constitute promising tools to uncoverand dissect the links between development and evolution, not onlyby comparing human and mouse PSC-derived systems, but also byimplementing reprogramming of somatic cells of other mammalianspecies in which experimental manipulation in embryos is difficult,such as non-human primates (Wunderlich et al., 2014). This mightshed further light on the evolutionary conservation and divergenceof various other aspects of brain development, such as the gyrationpatterns of the cerebral cortex that have been independently acquiredin multiple mammalian lineages (Lui et al., 2011).

From a molecular perspective, a number of genes have beenidentified recently, mostly through transcriptomics/genomicsapproaches, that may be involved in human-specific features of brain

1 month

B Mouse

ACortical

differentiation

Mouse or human PSCs PSC-derivedcortical cell

Dissociation Transplantation

Cortex

3 weeks 2 months 6 months 9 months

C Human

Fig. 5. Modeling neuronal maturation usingxenotransplantation. Cortical pyramidal neuronsderived from mouse or human PSCs displayspecies-specific maturation followingtransplantation in the mouse neonatal cortex (A).Whereas a mature-like pattern can be observed1 month post-transplantation with mouse ESC-derived cortical cells (B), human cells reach a similarstage 9 months post-transplantation (C). Scalebars: 20 µm. Images are adapted from Espuny-Camacho et al. (2013) and Gaspard et al. (2008).

3146

REVIEW Development (2015) 142, 3138-3150 doi:10.1242/dev.120568

DEVELO

PM

ENT

MouseorhumanESC

ESC-derivedcorticalneuronsandastrocytes

demonstrate the possibility of using PSC-based in vitro models forstudying synaptic development in the human, although much workwill be necessary to relate in vitro findings to in vivo situations. Inthis context, two lines of investigations should probably bedeveloped further to assess the connectivity patterns of humanneurons in a more physiologically relevant way. On the one hand, itwill be interesting to use 3Dmodels of corticogenesis, or other brainstructures, to dissect precise patterns of synaptic connectivity(Kadoshima et al., 2013; Lancaster et al., 2013). On the other hand,the transplantation of human PSC-derived cells into recipient mice,which enables human neuron development and connectivity to befollowed for up to one year (Espuny-Camacho et al., 2013), couldconstitute a promising tool to study the impact of genes orenvironment on human neuron connectivity in an in vivo contextthat is experimentally tractable.

Conclusions, challenges and future perspectivesWhile it has become trivial to discuss the potential of PSCtechnology for translational purposes, such as disease modeling andcell therapy, their potential usefulness as a tool in basicdevelopmental biology had remained less clear. From the datadiscussed here, PSC-derivedmodels of neural development have thepotential to do much more than just extend the findings previouslyobtained in animal models to the human setting. In particular, PSCmodels have revealed, in sometimes striking ways, the extent to

which cell diversity and spatial and temporal patterning can berobustly recapitulated in simplistic in vitro settings, emphasizing theimportance of intrinsic self-organization during development. Eventhough this could be considered trivial (after all, the embryo itself isobviously capable of self-organization), the versatility andaccessibility of PSC models makes them ideal to study theunderlying mechanisms, in particular for those structures that arespecific to organisms developing in utero. This should constitute auseful source of inspiration to design new experiments in vitro, aswell in in vivo models, to understand the underlying mechanisms.

Another striking observation is that PSC models recapitulatefaithfully the timing of developmental events, and in a species-specific way. PSCs thereby constitute promising tools to uncoverand dissect the links between development and evolution, not onlyby comparing human and mouse PSC-derived systems, but also byimplementing reprogramming of somatic cells of other mammalianspecies in which experimental manipulation in embryos is difficult,such as non-human primates (Wunderlich et al., 2014). This mightshed further light on the evolutionary conservation and divergenceof various other aspects of brain development, such as the gyrationpatterns of the cerebral cortex that have been independently acquiredin multiple mammalian lineages (Lui et al., 2011).

From a molecular perspective, a number of genes have beenidentified recently, mostly through transcriptomics/genomicsapproaches, that may be involved in human-specific features of brain

1 month

B Mouse

ACortical

differentiation

Mouse or human PSCs PSC-derivedcortical cell

Dissociation Transplantation

Cortex

3 weeks 2 months 6 months 9 months

C Human

Fig. 5. Modeling neuronal maturation usingxenotransplantation. Cortical pyramidal neuronsderived from mouse or human PSCs displayspecies-specific maturation followingtransplantation in the mouse neonatal cortex (A).Whereas a mature-like pattern can be observed1 month post-transplantation with mouse ESC-derived cortical cells (B), human cells reach a similarstage 9 months post-transplantation (C). Scalebars: 20 µm. Images are adapted from Espuny-Camacho et al. (2013) and Gaspard et al. (2008).

3146

REVIEW Development (2015) 142, 3138-3150 doi:10.1242/dev.120568

DEVELO

PM

ENT

demonstrate the possibility of using PSC-based in vitro models forstudying synaptic development in the human, although much workwill be necessary to relate in vitro findings to in vivo situations. Inthis context, two lines of investigations should probably bedeveloped further to assess the connectivity patterns of humanneurons in a more physiologically relevant way. On the one hand, itwill be interesting to use 3Dmodels of corticogenesis, or other brainstructures, to dissect precise patterns of synaptic connectivity(Kadoshima et al., 2013; Lancaster et al., 2013). On the other hand,the transplantation of human PSC-derived cells into recipient mice,which enables human neuron development and connectivity to befollowed for up to one year (Espuny-Camacho et al., 2013), couldconstitute a promising tool to study the impact of genes orenvironment on human neuron connectivity in an in vivo contextthat is experimentally tractable.

Conclusions, challenges and future perspectivesWhile it has become trivial to discuss the potential of PSCtechnology for translational purposes, such as disease modeling andcell therapy, their potential usefulness as a tool in basicdevelopmental biology had remained less clear. From the datadiscussed here, PSC-derivedmodels of neural development have thepotential to do much more than just extend the findings previouslyobtained in animal models to the human setting. In particular, PSCmodels have revealed, in sometimes striking ways, the extent to

which cell diversity and spatial and temporal patterning can berobustly recapitulated in simplistic in vitro settings, emphasizing theimportance of intrinsic self-organization during development. Eventhough this could be considered trivial (after all, the embryo itself isobviously capable of self-organization), the versatility andaccessibility of PSC models makes them ideal to study theunderlying mechanisms, in particular for those structures that arespecific to organisms developing in utero. This should constitute auseful source of inspiration to design new experiments in vitro, aswell in in vivo models, to understand the underlying mechanisms.

Another striking observation is that PSC models recapitulatefaithfully the timing of developmental events, and in a species-specific way. PSCs thereby constitute promising tools to uncoverand dissect the links between development and evolution, not onlyby comparing human and mouse PSC-derived systems, but also byimplementing reprogramming of somatic cells of other mammalianspecies in which experimental manipulation in embryos is difficult,such as non-human primates (Wunderlich et al., 2014). This mightshed further light on the evolutionary conservation and divergenceof various other aspects of brain development, such as the gyrationpatterns of the cerebral cortex that have been independently acquiredin multiple mammalian lineages (Lui et al., 2011).

From a molecular perspective, a number of genes have beenidentified recently, mostly through transcriptomics/genomicsapproaches, that may be involved in human-specific features of brain

1 month

B Mouse

ACortical

differentiation

Mouse or human PSCs PSC-derivedcortical cell

Dissociation Transplantation

Cortex

3 weeks 2 months 6 months 9 months

C Human

Fig. 5. Modeling neuronal maturation usingxenotransplantation. Cortical pyramidal neuronsderived from mouse or human PSCs displayspecies-specific maturation followingtransplantation in the mouse neonatal cortex (A).Whereas a mature-like pattern can be observed1 month post-transplantation with mouse ESC-derived cortical cells (B), human cells reach a similarstage 9 months post-transplantation (C). Scalebars: 20 µm. Images are adapted from Espuny-Camacho et al. (2013) and Gaspard et al. (2008).

3146

REVIEW Development (2015) 142, 3138-3150 doi:10.1242/dev.120568

DEVELOPM

ENT

demonstrate the possibility of using PSC-based in vitro models forstudying synaptic development in the human, although much workwill be necessary to relate in vitro findings to in vivo situations. Inthis context, two lines of investigations should probably bedeveloped further to assess the connectivity patterns of humanneurons in a more physiologically relevant way. On the one hand, itwill be interesting to use 3Dmodels of corticogenesis, or other brainstructures, to dissect precise patterns of synaptic connectivity(Kadoshima et al., 2013; Lancaster et al., 2013). On the other hand,the transplantation of human PSC-derived cells into recipient mice,which enables human neuron development and connectivity to befollowed for up to one year (Espuny-Camacho et al., 2013), couldconstitute a promising tool to study the impact of genes orenvironment on human neuron connectivity in an in vivo contextthat is experimentally tractable.

Conclusions, challenges and future perspectivesWhile it has become trivial to discuss the potential of PSCtechnology for translational purposes, such as disease modeling andcell therapy, their potential usefulness as a tool in basicdevelopmental biology had remained less clear. From the datadiscussed here, PSC-derivedmodels of neural development have thepotential to do much more than just extend the findings previouslyobtained in animal models to the human setting. In particular, PSCmodels have revealed, in sometimes striking ways, the extent to

which cell diversity and spatial and temporal patterning can berobustly recapitulated in simplistic in vitro settings, emphasizing theimportance of intrinsic self-organization during development. Eventhough this could be considered trivial (after all, the embryo itself isobviously capable of self-organization), the versatility andaccessibility of PSC models makes them ideal to study theunderlying mechanisms, in particular for those structures that arespecific to organisms developing in utero. This should constitute auseful source of inspiration to design new experiments in vitro, aswell in in vivo models, to understand the underlying mechanisms.

Another striking observation is that PSC models recapitulatefaithfully the timing of developmental events, and in a species-specific way. PSCs thereby constitute promising tools to uncoverand dissect the links between development and evolution, not onlyby comparing human and mouse PSC-derived systems, but also byimplementing reprogramming of somatic cells of other mammalianspecies in which experimental manipulation in embryos is difficult,such as non-human primates (Wunderlich et al., 2014). This mightshed further light on the evolutionary conservation and divergenceof various other aspects of brain development, such as the gyrationpatterns of the cerebral cortex that have been independently acquiredin multiple mammalian lineages (Lui et al., 2011).

From a molecular perspective, a number of genes have beenidentified recently, mostly through transcriptomics/genomicsapproaches, that may be involved in human-specific features of brain

1 month

B Mouse

ACortical

differentiation

Mouse or human PSCs PSC-derivedcortical cell

Dissociation Transplantation

Cortex

3 weeks 2 months 6 months 9 months

C Human

Fig. 5. Modeling neuronal maturation usingxenotransplantation. Cortical pyramidal neuronsderived from mouse or human PSCs displayspecies-specific maturation followingtransplantation in the mouse neonatal cortex (A).Whereas a mature-like pattern can be observed1 month post-transplantation with mouse ESC-derived cortical cells (B), human cells reach a similarstage 9 months post-transplantation (C). Scalebars: 20 µm. Images are adapted from Espuny-Camacho et al. (2013) and Gaspard et al. (2008).

3146

REVIEW Development (2015) 142, 3138-3150 doi:10.1242/dev.120568

DEVELO

PM

ENT

DissociationCortical

differentiation

Transplant

Newborn

Adult

B

Gene variant function in brain homeostasis

Introduce ancient gene variants into modern human genes in hESC

Archaic humanModern human

In vitro In vivo

iPSC transplantGenetic mouse models

Pathological

GFAP GFAP/tdTOMATO

GFP

/GFA

P/ho

st v

esse

l (td

TOM

)

Neurons, astrocytes and blood vessels inside the graft

Guillermo Oliver, Northwestern University

Pierre Vanderhaeghen,Universite Libre de Bruxelles

Jaume Mora and Cinzia Lavarino, HSJD

Collaborators

Eske Willerslev, U. CopenhaghenDaniel Geschwind, UCLAJudith Armstrong, HSJD

Jurgen Knoblich, IMBA Vienna

Miguel Manzanares, CBM, Madrid

Holger Heyn, CNAG-CRG

Oscar Lao, CNAG-CRG

Tomàs Marquès, IBE-UPF

Roderic Guigó, CRG

Jaume Bertranpetit